1. 8: Principles of Reliable Data Transfer
Last Modified:
4/7/2019 1:15:15 PM
Slides adapted from:
J.F Kurose and K.W. Ross,
3: Transport Layer

2. Roadmap UDP is a very thin layer over IP
multiplexing /demultiplexing
error detection
TCP does these things also and then adds some other significant features
TCP is quite a bit more complicated and subtle
We are not going to jump right into TCP
Start gently thinking about principles of reliable message transfer in general
3: Transport Layer

3. The Problem
Problem: send big message (broken into pieces) over unreliable channel such that it arrives on other side in its entirety and in the right order
No out of band communication! All communication sent along with the pieces of the message
Receiver allowed to send information back but only over the same unreliable channel!
3: Transport Layer

4. Intuition: Faxing a document With Flaky Machine
Can’t talk to person on the other side any other way
Number the pages – so sender can put back together
Let receiver send you a fax back saying what pages they have and what they still need (include your fax number on the document!)
What if the receiver sends their responses with a flaky fax machine too?
What if it is a really big document? No point in overwhelming the receiver. Receiver might like to be able to tell you send first 10 pages then 10 more…
How does receiver know when they have it all? Special last page? Cover sheet that said how many to expect?
3: Transport Layer

5. Principles of Reliable data transfer
Solving this problem is one on top-10 list of most important networking topics!
important in application, transport, link layers
Characteristics of unreliable channel will determine complexity of reliable data transfer protocol– what is worst underlying channel can do?
Drop packets/pages?
Corrupt packet/pages (even special ones like the cover sheet or the receiver’s answer?)
Reorder packets/pages?
Corrupt (checksum catches), drop, reorder (delay and reinject at random times)
3: Transport Layer

6. Reliable data transfer: getting started
rdt_send(): called from above, (e.g., by app.). Passed data to
deliver to receiver upper layer
deliver_data(): called by rdt to deliver data to upper
send
side
receive
side
udt_send(): called by rdt,
to transfer packet over
unreliable channel to receiver
rdt_rcv(): called when packet arrives on rcv-side of channel
3: Transport Layer

7. Reliable data transfer: getting started
We’ll:
incrementally develop sender, receiver sides of reliable data transfer protocol (rdt)
consider only unidirectional data transfer
but control info will flow on both directions!
use finite state machines (FSM) to specify sender, receiver
event causing state transition
actions taken on state transition
state 1
state: when in this “state” next state uniquely determined by next event
state 2
event
actions
3: Transport Layer

8. Rdt1.0: reliable transfer over a reliable channel
underlying channel perfectly reliable (so this should be easy )
no bit errors
no loss of packets
separate FSMs for sender, receiver:
sender sends data into underlying channel
receiver read data from underlying channel
Illustrate – one person is application, one transport layer, one network, other side transport layer, other side application
Person takes letter – gives letter plus says who to send it to to transport layer
transport layer puts it in an envelope with address of who to send it to on
Outside – gives it to network – in this case network perfectly reliable
Hands it over - ….
3: Transport Layer

9. Rdt2.0: channel with bit errors
underlying channel may flip bits in packet
checksum to detect bit errors
the question: how to recover from errors:
acknowledgements (ACKs): receiver explicitly tells sender that pkt received OK
negative acknowledgements (NAKs): receiver explicitly tells sender that pkt had errors
sender retransmits pkt on receipt of NAK
new mechanisms in rdt2.0 (beyond rdt1.0):
error detection
receiver feedback: control msgs (ACK,NAK) rcvr->sender
How do humans recover from “errors”
during conversation?
Transport Layer

10. Rdt2.0: channel with bit errors
underlying channel may flip bits in packet (can’t drop or reorder packets)
recall: UDP checksum to detect bit errors
Once can have problems, the receiver must give the sender feedback (either that or the sender would just have to keep sending copy after copy forever to be sure)
After receiving a packet, the receiver could say one of two things:
acknowledgements (ACKs): receiver explicitly tells sender that pkt received OK
negative acknowledgements (NAKs): receiver explicitly tells sender that pkt had errors
sender retransmits pkt on receipt of NAK
human scenarios using ACKs, NAKs?
Try scenario before if network can change letters – won’t work
Ok how do we fix it!!
3: Transport Layer

11. Rdt2.0: channel with bit errors
new mechanisms in rdt2.0 (beyond rdt1.0):
receiver feedback: control msgs (ACK,NAK) rcvr->sender (let receiver fax you back info?)
Possible retransmission – detection of duplicates (number fax pages?)
error detection (checksums? Cover sheet summary?)
Try scenario before if network can change letters – won’t work
Ok how do we fix it!!
3: Transport Layer

12. rdt2.0: FSM specification
Now need a checksum to summarize info (letter between r and z)
Plus now info will flow in other direction – write on envelopes (or colored paper)
ACK or NACK
Change ACK or NACK to FLACK
What should sender do?? Assume ok and continue..but what if NACK
Assume wrong and retransmit .. But what if ACK
sender FSM
receiver FSM
3: Transport Layer

13. rdt2.0: in action (no errors)
sender FSM
receiver FSM
3: Transport Layer

14. rdt2.0: in action (error scenario)
sender FSM
receiver FSM
3: Transport Layer

15. rdt2.0 has a fatal flaw! What happens if ACK/NAK corrupted?
sender doesn’t know what happened at receiver!
FSM implied could tell if it was and ACK or a NACK
What if is a FLACK?
What to do?
Assume it was an ACK and transmit next? What if it was a NACK? Missing data
Assume it was a NACK and retransmit; What if it was an ACK? Duplicate data
Handling duplicates:
To detect duplicate, sender adds sequence number to each pkt
sender retransmits current pkt if ACK/NAK garbled
If receiver has pkt with that number already it will discards (I.e. not deliver up duplicate pkt)
3: Transport Layer

16. rdt2.1: sender, handles garbled ACK/NAKs
New:
compute_chksum
corrupt()
3: Transport Layer

17. rdt2.1: receiver, handles garbled ACK/NAKs
If not corrupt, always
send ACK, but only
Deliver_data first time
3: Transport Layer

18. rdt2.1: discussion Receiver: Sender:
must check if received packet is duplicate
state indicates whether 0 or 1 is expected pkt seq #
Note: This protocol can also handle if the channel can duplicate packets
note: when can sender and receiver safely exit? receiver can not know if its last ACK/NAK received OK at sender
Missing connection termination procedure
Sender:
seq # added to pkt
two seq. #’s (0,1) will suffice. Why?
must check if received ACK/NAK corrupted
twice as many states
state must “remember” whether “current” pkt has 0 or 1 seq. #
3: Transport Layer

19. rdt2.2: a NAK-free protocol
Less intuitive but getting us closer to TCP
same functionality as rdt2.1, using NAKs only
instead of NAK, receiver sends ACK for last pkt received OK (or for other number on the first receive)
receiver must explicitly include seq # of pkt being ACKed
duplicate (or unexpected) ACK at sender results in same action as NAK: retransmit current pkt
TCP really ACKS the next thing it wants
3: Transport Layer

20. rdt2.2: sender, receiver fragments

21. rdt3.0: channels with errors (and duplicates) and loss
New assumption: underlying channel can also lose packets (data or ACKs)
How to deal with loss? Retransmission plus seq # to detect duplicates
but not enough
Q: how to detect loss?
Approach: sender waits “reasonable” amount of time for ACK
retransmits if no ACK received in this time
if pkt (or ACK) just delayed (not lost):
retransmission will be duplicate, but use of seq. #’s already handles this
receiver must specify seq # of pkt being ACKed
requires countdown timer
3: Transport Layer

22. rdt3.0 sender
Start_timer
Timeout events
3: Transport Layer

23. rdt3.0 in action
3: Transport Layer

24. rdt3.0 in action
3: Transport Layer

25. Stop and Wait Rdt3.0 also called Stop and Wait
Sender sends one packet, then waits for receiver response
What is wrong with stop and wait?
Slow!! Must wait full round trip time between each send
Obvious Fix?
Instead send lots, then stop and wait
Call this a pipelined protocol because many packets in the pipeline at the same time
3: Transport Layer